[PATCH v2] drm/panthor: Reset queue slots if termination fails

[PATCH v2] drm/panthor: Reset queue slots if termination fails

[Ashley Smith]

This fixes a bug where if we timeout after a suspend and the termination
fails, due to waiting on a fence that will never be signalled for
example, we do not resume the group correctly. The fix forces a reset
for groups that are not terminated correctly.

Signed-off-by: Ashley Smith <ashley.smith@collabora.com>
---
Changes in v2:
 - Fixed syntax error
---
 drivers/gpu/drm/panthor/panthor_sched.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/panthor/panthor_sched.c b/drivers/gpu/drm/panthor/panthor_sched.c
index 43ee57728de5..65d8ae3dcac1 100644
--- a/drivers/gpu/drm/panthor/panthor_sched.c
+++ b/drivers/gpu/drm/panthor/panthor_sched.c
@@ -2727,8 +2727,17 @@ void panthor_sched_suspend(struct panthor_device *ptdev)
 			 * automatically terminate all active groups, so let's
 			 * force the state to halted here.
 			 */
-			if (csg_slot->group->state != PANTHOR_CS_GROUP_TERMINATED)
+			if (csg_slot->group->state != PANTHOR_CS_GROUP_TERMINATED) {
 				csg_slot->group->state = PANTHOR_CS_GROUP_TERMINATED;
+
+				/* Reset the queue slots manually if the termination
+				 * request failed.
+				 */
+				for (i = 0; i < group->queue_count; i++) {
+					if (group->queues[i])
+						cs_slot_reset_locked(ptdev, csg_id, i);
+				}
+			}
 			slot_mask &= ~BIT(csg_id);
 		}
 	}

base-commit: 9934ab18051118385c7ea44d8e14175edbe6dc9c
-- 
2.43.0

Re: [PATCH v2] drm/panthor: Reset queue slots if termination fails

[Liviu Dudau]

On Mon, May 19, 2025 at 03:50:19PM +0100, Ashley Smith wrote:
> This fixes a bug where if we timeout after a suspend and the termination
> fails, due to waiting on a fence that will never be signalled for
> example, we do not resume the group correctly. The fix forces a reset
> for groups that are not terminated correctly.
> 
> Signed-off-by: Ashley Smith <ashley.smith@collabora.com>

Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>

Best regards,
Liviu

> ---
> Changes in v2:
>  - Fixed syntax error
> ---
>  drivers/gpu/drm/panthor/panthor_sched.c | 11 ++++++++++-
>  1 file changed, 10 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/gpu/drm/panthor/panthor_sched.c b/drivers/gpu/drm/panthor/panthor_sched.c
> index 43ee57728de5..65d8ae3dcac1 100644
> --- a/drivers/gpu/drm/panthor/panthor_sched.c
> +++ b/drivers/gpu/drm/panthor/panthor_sched.c
> @@ -2727,8 +2727,17 @@ void panthor_sched_suspend(struct panthor_device *ptdev)
>  			 * automatically terminate all active groups, so let's
>  			 * force the state to halted here.
>  			 */
> -			if (csg_slot->group->state != PANTHOR_CS_GROUP_TERMINATED)
> +			if (csg_slot->group->state != PANTHOR_CS_GROUP_TERMINATED) {
>  				csg_slot->group->state = PANTHOR_CS_GROUP_TERMINATED;
> +
> +				/* Reset the queue slots manually if the termination
> +				 * request failed.
> +				 */
> +				for (i = 0; i < group->queue_count; i++) {
> +					if (group->queues[i])
> +						cs_slot_reset_locked(ptdev, csg_id, i);
> +				}
> +			}
>  			slot_mask &= ~BIT(csg_id);
>  		}
>  	}
> 
> base-commit: 9934ab18051118385c7ea44d8e14175edbe6dc9c
> -- 
> 2.43.0
> 

-- 
====================
| I would like to |
| fix the world,  |
| but they're not |
| giving me the   |
 \ source code!  /
  ---------------
    ¯\_(ツ)_/¯

Re: [PATCH v2] drm/panthor: Reset queue slots if termination fails

[Steven Price]

On 19/05/2025 15:50, Ashley Smith wrote:
> This fixes a bug where if we timeout after a suspend and the termination
> fails, due to waiting on a fence that will never be signalled for
> example, we do not resume the group correctly. The fix forces a reset
> for groups that are not terminated correctly.
> 
> Signed-off-by: Ashley Smith <ashley.smith@collabora.com>

Reviewed-by: Steven Price <steven.price@arm.com>

> ---
> Changes in v2:
>  - Fixed syntax error
> ---
>  drivers/gpu/drm/panthor/panthor_sched.c | 11 ++++++++++-
>  1 file changed, 10 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/gpu/drm/panthor/panthor_sched.c b/drivers/gpu/drm/panthor/panthor_sched.c
> index 43ee57728de5..65d8ae3dcac1 100644
> --- a/drivers/gpu/drm/panthor/panthor_sched.c
> +++ b/drivers/gpu/drm/panthor/panthor_sched.c
> @@ -2727,8 +2727,17 @@ void panthor_sched_suspend(struct panthor_device *ptdev)
>  			 * automatically terminate all active groups, so let's
>  			 * force the state to halted here.
>  			 */
> -			if (csg_slot->group->state != PANTHOR_CS_GROUP_TERMINATED)
> +			if (csg_slot->group->state != PANTHOR_CS_GROUP_TERMINATED) {
>  				csg_slot->group->state = PANTHOR_CS_GROUP_TERMINATED;
> +
> +				/* Reset the queue slots manually if the termination
> +				 * request failed.
> +				 */
> +				for (i = 0; i < group->queue_count; i++) {
> +					if (group->queues[i])
> +						cs_slot_reset_locked(ptdev, csg_id, i);
> +				}
> +			}
>  			slot_mask &= ~BIT(csg_id);
>  		}
>  	}
> 
> base-commit: 9934ab18051118385c7ea44d8e14175edbe6dc9c

Re: [PATCH v2] drm/panthor: Reset queue slots if termination fails

[Boris Brezillon]

On Mon, 19 May 2025 15:50:19 +0100
Ashley Smith <ashley.smith@collabora.com> wrote:

> This fixes a bug where if we timeout after a suspend and the termination
> fails, due to waiting on a fence that will never be signalled for
> example, we do not resume the group correctly. The fix forces a reset
> for groups that are not terminated correctly.
> 
> Signed-off-by: Ashley Smith <ashley.smith@collabora.com>

We clearly need a Fixes tag, otherwise "drm/panthor: Make the timeout
per-queue instead of per-job" will be backported, but not this one,
which will lead to UAF. Can you send a v5 with both patches in the same
series, each with a proper Fixes tag?

> ---
> Changes in v2:
>  - Fixed syntax error
> ---
>  drivers/gpu/drm/panthor/panthor_sched.c | 11 ++++++++++-
>  1 file changed, 10 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/gpu/drm/panthor/panthor_sched.c b/drivers/gpu/drm/panthor/panthor_sched.c
> index 43ee57728de5..65d8ae3dcac1 100644
> --- a/drivers/gpu/drm/panthor/panthor_sched.c
> +++ b/drivers/gpu/drm/panthor/panthor_sched.c
> @@ -2727,8 +2727,17 @@ void panthor_sched_suspend(struct panthor_device *ptdev)
>  			 * automatically terminate all active groups, so let's
>  			 * force the state to halted here.
>  			 */
> -			if (csg_slot->group->state != PANTHOR_CS_GROUP_TERMINATED)
> +			if (csg_slot->group->state != PANTHOR_CS_GROUP_TERMINATED) {
>  				csg_slot->group->state = PANTHOR_CS_GROUP_TERMINATED;
> +
> +				/* Reset the queue slots manually if the termination
> +				 * request failed.
> +				 */
> +				for (i = 0; i < group->queue_count; i++) {
> +					if (group->queues[i])
> +						cs_slot_reset_locked(ptdev, csg_id, i);
> +				}
> +			}
>  			slot_mask &= ~BIT(csg_id);
>  		}
>  	}
> 
> base-commit: 9934ab18051118385c7ea44d8e14175edbe6dc9c